We have a network that comprises a bunch of Windows Servers, Linux servers, CentOS, and a variety of network devices, such as Cisco routers, Cisco switches, Riverbeds, and some VeloCloud. We use the service to monitor and alert us to any potential issues that we may be having. We also use it to do some pink tests and to monitor the availability of some websites as well.
The whole purpose is to give the IT team a heads-up, before the user base is aware of an issue. There are different levels in the system from "warning" up to "critical" that can let us see that a situation might be developing, that we might be having a problem with a system. We can proactively take care of it before it transitions to a level where it might affect our users and prevent them from doing their daily work.